The present invention relates to television receivers for displaying/outputting video images such as television broadcasting signals and, more particularly, a television receiver configured to recognize voices of a user and control the receiver body based on the voices.
The voice recognition technology for recognizing voices inputted includes the technique of extracting a word-like portion of a voice being uttered, which is so-called word spotting, the technique of recognizing successive voices by using HMM (Hidden Marcov Model), and like techniques.
Conventionally widespread television receivers are generally provided with an infrared wireless remote controller for remote control of such a television receiver. Existing television receivers of another type adopt the aforementioned voice recognition technology for remote control by voices.
For example, when the user utters a voice representing a command to power ON/OFF, change channels or adjust the sound volume or a like command, the television receiver receives the voice by means of a microphone or the like and recognizes the voice. This conventional television receiver executes the command to power ON/OFF, change channels or adjust the sound volume or the like command according to the voice recognized.
A remote controller has been proposed as disclosed in Japanese Patent Laid-Open Publication No. 2001-275176. This remote controller includes a microphone for receiving voices uttered by the user, a voice recognition section for recognizing these voices, and a voice recognition switch for enabling or disabling the voice recognition operation of the voice recognition section switchably. The remote controller is configured to recognize voices uttered by the user only when the voice recognition switch is in the depressed position and then transmit a command (control code) for remote-controlling a controlled apparatus such as a television receiver according to the voices.
The aforementioned conventional television receiver, however, is configured to receive voices by means of the microphone and, hence, sometimes receives voice and sound of a TV program along with voices uttered by the user undesirably. Accordingly, the conventional television receiver often experiences malfunctions because the television receiver recognizes a voice that does not meet the intention of the user and then executes a command according to such a voice.
The aforementioned remote controller disclosed in Japanese Patent Laid-Open Publication No. 2001-275176 recognizes voices uttered by the user only when the voice recognition switch is in the depressed position. Depressing of the switch relies upon a manual operation performed by the user. This means that the user has to perform a manual operation every time a command is given to power ON/OFF, change channels, adjust the sound volume, or the like. For this reason, the user feels it inconvenience. In addition, such an inconvenience manual operation by the user does not meet the purpose of “remote-controlling the television receiver by voice” and, hence, this remote controller is, in no way, different from an infrared wireless remote controller.
Accordingly, a feature of the present invention is to provide a television receiver which does not need any manual operation by the user and which has a lowered possibility of a malfunction caused by voice and sound of a television program.